DEVELOPMENT OF A MULTI-BIOMARKER SERUM SIGNATURE FOR BREAST CANCER DETECTION USING ANGIOGENIC, INFLAMMATORY, AND TISSUE REMODELLING BIOMARKERS

Abstract

Background: Breast cancer grows by forming new blood vessels through a process called angiogenesis. Single blood biomarker often do not detect breast cancer accurately. This study examined a panel of biomarkers linked to angiogenesis, inflammation, and tissue remodeling to identify a reliable blood-based signature for breast cancer detection. Methods: Serum samples from 55 women with breast cancer and 80 healthy controls were analyzed using the MAGPIX xMAP multiplex system. 20 circulating biomarkers were measured. Differences between the two groups were tested using the Mann–Whitney U test, with Benjamini–Hochberg false discovery rate (FDR) correction for multiple comparisons. Receiver operating characteristic (ROC) analysis assessed the performance of each biomarker. Logistic regression and random forest models were used to develop prediction models and rank the importance of each biomarker. Results: After FDR correction, 12 biomarkers remained significantly associated with breast cancer. Most of these biomarkers, including osteopontin, sNeuropilin-1, suPAR, sVEGFR2, sVEGFR1, sc-Kit/SCFR, sHER2, Tenascin-C, sIL-6Ra, and sPECAM-1, were lower in patients with breast cancer, while sTie-2 showed the greatest increase. Osteopontin was the best single biomarker (AUC = 0.702; q = 0.0011), with 82% lower level than in healthy controls. A five-biomarker panel achieved excellent diagnostic accuracy (AUC ≈ 0.90), which improved to about 0.93 when age was included. Conclusion: A panel of five blood biomarkers showed strong ability to identify breast cancer. This simple, non invasive approach may improve breast cancer detection, but larger studies are needed before it can be used in routine clinical practice.